The table
Half-life describes the time for plasma concentration to fall by half during the relevant elimination phase. The table is an approximate educational comparison, not product-label instructions. The selected bibliography supports only the compounds and mechanisms identified in its scope notes; it does not verify every table entry. Individual values and the relationship between half-life and steady state depend on formulation, absorption and study conditions.
| Peptide | Mechanism | Half-life | Dosing | How the half-life is achieved |
|---|---|---|---|---|
| Native GLP-1 | GLP-1 hormone | ≈ 1–2 min | — | None: DPP-4 cleavage + renal clearance |
| Pramlintide | Amylin analogue | ≈ 48 min | Before meals | Proline substitutions only; no lipid |
| Exenatide | Exendin-4 (GLP-1R) | ≈ 2.4 h | Twice daily | Gly2 resists DPP-4; renal clearance limits it |
| Lixisenatide | Exendin-4 analogue | ≈ 3 h | Once daily | C-terminal lysine tail; renal clearance |
| Liraglutide | GLP-1 analogue | ≈ 13 h | Once daily | C16 palmitate on Lys26 → albumin binding |
| Dulaglutide | GLP-1–Fc fusion | ≈ 5 days | Once weekly | Fusion to IgG4 Fc → FcRn recycling |
| Tirzepatide | GIP/GLP-1 dual | ≈ 5 days (117 h) | Once weekly | Aib2 + C20 diacid on Lys20 |
| Retatrutide | GIP/GLP-1/GCG triple | ≈ 6 days | Once weekly | Aib2 + C20 diacid on Lys17 |
| Semaglutide | GLP-1 analogue | ≈ 7 days (145–165 h) | Once weekly | Aib8 + C18 diacid on Lys26 |
| Cagrilintide | Amylin analogue | ≈ 7–8 days (159–195 h) | Once weekly | Proline substitutions + C20 diacid on N-terminus |
| Survodutide | GCG/GLP-1 dual | Several days | Once weekly | Aib2 + C18 diacid |
| Mazdutide | GCG/GLP-1 dual | Several days | Once weekly | Aib2 + C20 diacid |
| Orforglipron | Non-peptide GLP-1R agonist | ≈ 1–2 days | Once daily (oral) | Small molecule; hepatic metabolism |
Why native GLP-1 lasts two minutes
Two enzymes and one organ dispose of GLP-1 almost as fast as the gut releases it. Dipeptidyl peptidase-4 (DPP-4) removes the N-terminal His-Ala within a minute, leaving inactive GLP-1(9-36). Neutral endopeptidase cuts the mid-chain. The kidney filters what remains, because a 3.3 kDa peptide passes freely through the glomerulus. The result is a plasma half-life of one to two minutes and a hormone that works only as a local, meal-timed signal. Every analog is an answer to one or more of these three clearance routes.
Strategy 1: block DPP-4
The simplest fix is to make the His-Ala bond uncleavable. Exendin-4 does this naturally with glycine at position 2, which is why exenatide survives DPP-4 — but it is still filtered by the kidney, so its half-life is only 2.4 hours. Semaglutide, tirzepatide, retatrutide, survodutide and mazdutide all replace the residue after the N-terminal histidine or tyrosine with α-aminoisobutyric acid (Aib), whose extra methyl group DPP-4 cannot accommodate. DPP-4 resistance alone buys hours, not days; the days come from strategy 2.
Strategy 2: ride on albumin
Serum albumin has a half-life of about three weeks and is too large to be filtered. A peptide that spends more than 99% of its time bound to albumin inherits protection from both filtration and proteases. Liraglutide showed that a C16 fatty acid on a lysine side chain does this well enough for once-daily dosing (13 hours). Semaglutide's C18 fatty diacid — with a carboxylate at each end and a γ-glutamate plus two OEG spacers — binds albumin roughly three times more tightly and, combined with Aib8, reaches 7 days. Tirzepatide and retatrutide use a C20 diacid; cagrilintide puts its C20 diacid on the N-terminus. The pattern is consistent: longer diacid and better spacer, longer half-life.
Strategy 3: fuse to an antibody fragment
Dulaglutide takes a different route. Two DPP-4-resistant GLP-1 analogues are fused to a modified human IgG4 Fc region, producing a 60 kDa protein far too large for renal filtration that is recycled by the neonatal Fc receptor (FcRn) like an antibody. The result is a half-life of about five days without any lipid. Albiglutide (withdrawn) used albumin itself as the fusion partner. Fc fusion works, but the molecules are proteins rather than peptides, manufactured in cells rather than synthesised, and cannot be co-formulated as simply as lipidated peptides.
What the numbers mean in practice
A 7-day half-life means weekly injections keep concentrations within about a two-fold range between peak and trough. It also means steady state takes four to five weeks to reach, which is why dose escalation is spaced at four-week intervals and why the effect of a change is not fully visible for a month. After the last dose, semaglutide takes five to seven weeks to clear, tirzepatide about four, liraglutide two to three days — relevant to washout before surgery, to switching between agents, and to how long side effects persist. Orforglipron, the non-peptide, is at the other end: daily dosing and clearance within days.
The incretin & metabolic co-agonist class at a glance
| Peptide | Receptors | Length | Mass (Da) | Half-life | Status |
|---|---|---|---|---|---|
| GLP-1 | GLP-1R | 30 aa | ≈ 3,297.6 | ≈ 1–2 min (DPP-4 cleavage + renal clearance) | Endogenous hormone — not a product |
| Liraglutide | GLP-1R | 31 aa | ≈ 3,751.2 | ≈ 13 h (once-daily dosing) | Approved: Victoza (2010, T2D), Saxenda (2014, weight) |
| Semaglutide | GLP-1R | 31 aa | ≈ 4,113.6 | ≈ 7 days (≈ 145–165 h in clinical PK studies) | Approved: Ozempic (2017, T2D), Rybelsus (2019, oral T2D), Wegovy (2021, weight) |
| Tirzepatide | GIPR + GLP-1R | 39 aa | ≈ 4,813.5 | ≈ 5 days (≈ 117 h) | Approved: Mounjaro (2022, T2D), Zepbound (2023, weight; 2024, obstructive sleep apnoea) |
| Retatrutide | GIPR + GLP-1R + GCGR | 39 aa | ≈ 4,731 | ≈ 6 days | Investigational (phase 3) |
| Cagrilintide | Amylin / calcitonin | 37 aa | ≈ 4,409 | ≈ 7–8 days (159–195 h across doses in the phase 1b study) | Investigational (phase 3, as CagriSema with semaglutide) |
| Survodutide | GCGR + GLP-1R | 29 aa | ≈ 4,232 | Supports once-weekly dosing (published population estimates are on the order of several days) | Investigational (phase 3: SYNCHRONIZE obesity programme; MASH programme) |
| Mazdutide | GCGR + GLP-1R | 34 aa | ≈ 4,476 | Supports once-weekly dosing (published estimates are on the order of several days) | Approved in China (NMPA): chronic weight management (27 June 2025) and type 2 diabetes (2025) |
| Amycretin | GLP-1R + amylin | — | — | Supports once-weekly subcutaneous dosing | Investigational (phase 2 results published 2025; later-phase trials under way) |
Verified from PubChem, IUPHAR and the primary discovery papers cited on each profile. Half-lives are approximate population means from published pharmacokinetics.
Sources and what they cover
These primary studies, official records and standards cover the topics linked below. The scope notes distinguish laboratory findings, human trials and regulatory records. This is a selected bibliography, not a systematic review or verification of every sentence; absence of a study here does not prove that no study exists.
- Deacon et al. (1995) — GLP-1 degradation in human plasma
Original peptide metabolism work; native GLP-1 degradation does not describe every engineered analogue.
Relevant sections: Why native GLP-1 lasts two minutes
- Lau et al. (2015) — Discovery of the once-weekly GLP-1 analogue semaglutide
Original molecular design and preclinical characterization, including modifications for prolonged exposure.
Relevant sections: Strategy 1: block DPP-4 · Strategy 2: ride on albumin
- Knudsen et al. (2000) — Potent GLP-1 derivatives with once-daily pharmacokinetic properties
Original lipidated GLP-1 analogue design; clinical indication and efficacy require separate evidence.
Relevant sections: Strategy 2: ride on albumin
- Urva et al. (2022) — LY3437943/retatrutide phase 1b study
Early human pharmacokinetics and safety; limited sample and follow-up.
Relevant sections: The table · What the numbers mean in practice
- Enebo et al. (2021) — Cagrilintide with semaglutide, phase 1b
Early coadministration study; pharmacokinetics and tolerability do not prove long-term outcomes.
Relevant sections: The table · What the numbers mean in practice
FAQ
What is the half-life of semaglutide?+
About 7 days (145–165 hours in clinical PK studies). Steady state takes 4–5 weeks; washout after the last dose takes 5–7 weeks.
What is the half-life of tirzepatide?+
About 5 days (≈117 hours). Steady state is reached after about 4 weekly doses and washout takes roughly 4 weeks.
What is the half-life of retatrutide?+
About 6 days, based on phase 1 pharmacokinetics — slightly longer than tirzepatide, slightly shorter than semaglutide.
Why does native GLP-1 last only minutes?+
DPP-4 removes its first two residues within a minute, neutral endopeptidase cuts the middle, and the kidney filters the rest. Analogs block DPP-4 with Aib and evade filtration by binding albumin.
How long does it take to reach steady state?+
About four to five half-lives — roughly 4–5 weeks for the once-weekly peptides. That is why dose escalation steps are spaced four weeks apart.
Which GLP-1 drug has the longest half-life?+
Among approved products, semaglutide at about 7 days. Cagrilintide, an investigational amylin analogue, is slightly longer at 7–8 days.
Related peptide profiles
More guides
- Why Peptides Have Short Half-Lives (And How Engineers Fix It)
- GLP-1 Peptides Explained: GLP-1, Semaglutide, Tirzepatide, Retatrutide
- Semaglutide vs Tirzepatide: Structure, Receptors, Half-Life and the SURMOUNT-5 Head-to-Head
- Retatrutide vs Tirzepatide: Triple vs Dual Agonist — Structure, Half-Life, Trial Data